At first blush, “the domain of general internal medicine” may seem to be something of an oxymoron. General internal medicine (GIM) is not defined by an organ system or by specialized knowledge or skills in the way that most specialties are. It does not possess exclusive mastery of any technique, diagnosis or treatment, yet it professes mastery of care in certain circumstances. There will probably never be reports on the domains of ophthalmology, orthopedic surgery or cardiology, whose objects of concern are more self-evident.
Medicine did not always have specialties. Their initial development was driven by economic self-interest, later to be reinforced by technological advances and expanding knowledge.1 Economic forces can similarly cause specialties to atrophy, through exclusion from the market, burdensome workloads, inadequate recompense and the like. The decline in interest in GIM has been substantial.2 If it is not reversed, we face the prospect that GIM may be heading for the dustbin of history.
In this very challenging context, the Report by Larson et al.3 is an extremely valuable call to the profession for radical change. Long gone are the days in which we could conduct genteel rounds at the hospital bedside and assume that students would aspire to be “compleat internists.” Students are very concerned now about debt load and quality of life. Furthermore, they see internists who are clearly not happy.4 Students’ judgments that they should go elsewhere are hardly surprising. Yet, to those of us who know and love what GIM has to offer, this is a troubling turn of events.
The Report insists that GIM can be a big tent: one in which practice can encompass both breadth and depth. At the same time, it maintains that within that tent, there need to be much more clearly defined areas of focus, such as selected chronic diseases in which appropriately trained generalists can provide the majority of the care, expertise in measurement of quality and outcomes of care, and provision of preventive services.
Expertise often will center upon patients with multiple chronic diseases. In some settings, the internist still will choose to emphasize primary care of patients who are generally well. The Report speaks of the “pluripotent” internist who can provide care that is both broad and deep. While the discipline needs to incorporate that range of competencies, it is less clear that it will be possible for an individual to do so.
A few points in the Report merit particular attention and discussion. The Report is ambiguous as to whether the training that is needed to prepare internists for their future roles can be accomplished in three years. Certainly, any such training associated with certificates of added qualification will require more time. Additional training time, however, may pose difficulties for a field that already has problems attracting trainees, unless this can be tied to a parallel restructuring of reimbursement, which the Report advocates. That kind of systemic change will need to be pursued in the public sector and supported by other specialties. Second, the proposed roles of the internist include team leadership in the clinical setting, and in assuring the quality of care that is provided. Internists will need the cooperation of medical organizations in establishing the systems needed and these roles. Yet the larger profession will not welcome more generous pay for internists within a pot that is not growing, or expanded roles in care if they are skeptical of the internists’ contributions.
This leads to the urgent need for research. Hospitalists and geriatricians both established their specialties in part based on studies that demonstrated that they were improving care or reducing costs.5–7 Studies in some other clinical areas have suggested that specialist care results in superior outcomes8 although these may relate more to experience than certification.9 Rather than throw in the towel, generalists need to investigate in a serious and scholarly way the impact of the care that they provide. It may well be that patients are best off with joint generalist-specialist care in some instances and with specialist care in others, but there is no doubt in my mind that general internists provide equivalent or superior care in some situations, notably in patients with multiple chronic illnesses.
What are the parameters that characterize the patient who benefits from having a generalist guiding their care? What diseases does he/she have? What is his/her age range? What is the level of complexity of care? What are the dimensions of care that the generalist provides most and least effectively? What skills are needed to raise the general internists’ skill level in other dimensions to that of the specialist? What are the characteristics of a practice in which general internist care is as good as it gets, or otherwise? Conversely, what is missed when a generalist is not involved in the care of a patient with a particular disease or certain characteristics?
The Medical Outcomes Study examined some of these issues in the 1980s10,11 It is time to explore them in more depth today. General Internal Medicine should pursue these questions with no less vigor than other groups. We need to approach these inquiries with open minds and accept the possibility that we will not be able to be all things to all people. Expertise of the kind that comes from spending much of one's time on a particular problem will be needed in some situations. Still, it should be possible to identify areas in which our work is associated with superior outcomes or greater cost-effectiveness. Funders who care about cost, quality and outcomes should support such work.
The challenge in education is no less daunting. While our academic departments will not want us to hurtle toward extinction, the various specialty societies will not be altogether pleased at the notion of a refurbished internist playing increasing roles in the care of diseases that they regard as their domains. We will have to convince many different players that curriculum reform and certification reform along these lines makes sense, then work to produce the curriculum that will make all of this possible.
In the current issue of the Journal of General Internal Medicine are several articles that illuminate the challenges that we face. Golin et al. found that adherence counseling for patients with HIV disease is more common among infectious diseases specialists, but also among physicians who perceive that they have enough time, reimbursement, skill and office space to do the counseling.12 Kripalani et al. found that hospitalists and general medicine attendings were more effective teachers on the inpatient service than were subspecialists.13 Huang et al. found inpatient resource utilization to be comparable between a general medicine clinic and a diabetes clinic, but that disease severity was greater in the diabetes clinic.14 Donohoe observes that the movement to boutique medical practices of internal medicine, particularly in the context of academic medical centers that have broad missions to promote the public health, is ethically problematic.15
Let's be sure that, 10 or 15 years hence, the evidence and training systems are in place to make it possible to construct a health care system that optimally addresses the needs of all patients. That will not always mean handing the patient up the line to the person with the narrowest area of expertise. That will certainly involve creating systems that monitor the care that is provided. That will also require that doctors who are not narrowly construed will have access to the resources and comanagement opportunities that will serve their patients the best. Will there be alternative visions competing with the one that we will rationally construct? You bet. But if GIM can move forward collecting evidence and refining its vision accordingly, our perspective should be quite persuasive.—Martin F. Shapiro MD, PhD,Division of General Internal Medicine and Health Services Research, University of California, Los Angeles, Los Angeles, Calif.